TECHNOLOGY EXPLAINER
CAR-T Cells
— a medicine that collects, rewrites, expands and returns the patient's own T cells
CAR-T cell therapy takes the patient's own T cells, gives them the gene for an artificial receptor (a CAR) that recognises a marker on cancer cells, and puts them back into the body. Seven products are approved in the United States and five in Japan. Each is a medicine and, at the same time, a small manufacturing lot started up for one patient. This article reads the structure, manufacturing process and timelines set out in the prescribing information through the eyes of materials and process engineering.
- What a CAR-T cell is (the short version)
- How it works: the structure of the chimeric antigen receptor (CAR)
- Gene transfer: retroviruses and lentiviruses
- The manufacturing process: from collection to infusion (process diagram)
- How long manufacturing takes, and manufacturing failure
- A materials engineer's view (1): the materials inside the process
- Approved products (FDA and PMDA)
- The challenge of solid tumours
- Towards allogeneic CAR-T and CAR-NK
- A materials engineer's view (2): who can shorten the wait?
- The challenges in summary
- Glossary / References / Claim-to-source audit
Sourced = stated in published material, prescribing information or a peer-reviewed paper (link given)
Our calculation = a figure this article derived, with the assumptions spelled out
Not yet confirmed = at the research stage, with no confirmed approval or track record
Structural summaries and readings about materials and processes are marked separately as Commentary.
This is a technology explainer written from the angle of manufacturing and materials, not a treatment recommendation or medical advice. CAR-T cell therapy has serious side effects, including cytokine release syndrome and neurotoxicity, and the prescribing information carries a boxed warning. The article does not compare therapeutic effect between products. For any individual treatment, always consult a medical professional.
1. What a CAR-T cell is (the short version)
- What it is: an autologous T cell therapy in which the patient's own T cells are collected, genetically modified outside the body to express a chimeric antigen receptor (CAR), expanded, and returned to the patientSourced
- What it targets: the products approved in the United States target either CD19, found on the surface of B cells, or BCMA, found on the surface of myeloma cells and others. All approved uses are for blood cancersSourced
- What is distinctive about manufacturing: the product is shipped frozen in a patient-specific container, and in the pivotal trials the median time from collection to the product arriving was 15 to 35 daysSourced
The prescribing information for CAR-T cell therapies states the proportion of patients for whom manufacturing failed in the clinical trials. The range is under 1% to 18%Sourced. With an ordinary medicine, nobody explains the risk of a manufacturing failure to the patient. In this therapy, the yield of the manufacturing process translates directly into whether the patient can be treated at all (our commentary).
2. How it works: the structure of the chimeric antigen receptor (CAR)
T cells are immune cells that use their own receptors to find and attack, for example, cells infected by viruses. A CAR is an artificial receptor that joins the "recognition part" of an antibody to the "switch part" of a T cell. Because it links parts of different origin, it is called "chimeric" (our commentary).
NCI (the US National Cancer Institute) explains that a CAR spans the cell membrane; the part outside the cell is made from laboratory-made antibody fragments (domains), which determine binding to the target, and the part inside carries "signalling" and "co-stimulatory" domains that transmit signals on binding and cause the T cells to multiply in the bodySourced.
Some products use CD28 as the co-stimulatory domain (Yescarta, Tecartus) and others 4-1BB (Kymriah, Abecma, Carvykti, Breyanzi, Aucatzyl)Sourced. Breyanzi is distinctive in that CD8-positive and CD4-positive T cells are activated and transduced separately and given as a combination of two components (a defined composition)Sourced.
3. Gene transfer: retroviruses and lentiviruses
The CAR gene is delivered into T cells with a viral vector modified so that it cannot replicate. Approved products use two kinds: gammaretroviral retroviral vectors and lentiviral vectorsSourced. Both integrate the CAR gene into the T cell's genome, so the CAR is passed on when the cell divides (our commentary).
| Product | Target | Vector | Activation (as stated in the label) |
|---|---|---|---|
| Kymriah | CD19 | Lentivirus | Activated with beads coated with anti-CD3/CD28 antibodies |
| Yescarta | CD19 | Retrovirus | Anti-CD3 antibody and IL-2 |
| Tecartus | CD19 | Retrovirus | Anti-CD3 and anti-CD28 antibodies and IL-2. Includes a T cell enrichment step |
| Breyanzi | CD19 | Lentivirus | CD8-positive and CD4-positive T cells activated and transduced separately |
| Abecma | BCMA | Lentivirus | Anti-CD3 and anti-CD28 antibodies and IL-2 |
| Carvykti | BCMA | Lentivirus | (T cells enriched and transduced) |
| Aucatzyl | CD19 | Lentivirus | (T cells enriched, activated and transduced) |
All Sourced (FDA prescribing information [Refs. 1 to 7]). Entries in brackets are those for which the activation reagents could not be confirmed in the prescribing information.
Integration into the genome also raises safety questions. The prescribing information for Yescarta and others states in a boxed warning that T cell malignancies have occurred following treatment with BCMA- and CD19-directed genetically modified autologous T cell immunotherapies, and calls for lifelong monitoringSourced. The Abecma prescribing information also notes that an analysis of the lentiviral vector's genomic insertion sites was carried outSourced.
4. The manufacturing process: from collection to infusion (process diagram)
The DESCRIPTION section of each FDA prescribing information sets out a manufacturing process with much the same skeletonSourced. Taking Yescarta as the example: the patient's peripheral blood mononuclear cells are obtained by standard apheresis; the T cells are enriched and activated in the presence of anti-CD3 antibody and IL-2, then transduced with a replication-incompetent retroviral vector carrying the CAR gene. The transduced T cells are expanded in culture, washed, formulated into a suspension and cryopreserved. The product must pass a sterility test before it is shipped as a frozen suspension in a patient-specific infusion bagSourced.
| Step | What can be confirmed in the prescribing information (examples) |
|---|---|
| 1 Collection | Peripheral blood mononuclear cells collected by standard apheresis (all products) |
| 3 Activation | Kymriah: beads coated with anti-CD3/CD28 antibodies. Yescarta: anti-CD3 antibody and IL-2. Abecma and Tecartus: anti-CD3 and anti-CD28 antibodies and IL-2 |
| 4 Transduction | Replication-incompetent retroviral or lentiviral vector |
| 6 Formulation and freezing | DMSO concentration: Yescarta and Carvykti 5%, Kymriah 7.5%; Breyanzi 75% CryoStor CS10 (containing 7.5% DMSO) |
| 7 Release and shipping | Shipped after passing sterility testing. Yescarta and Tecartus in a liquid nitrogen dry shipper at or below −150 °C. Stored at or below −120 °C at the site (Yescarta, Kymriah, Carvykti) |
| 8 Receipt | Patient identifiers on the cassette and infusion bag checked. Yescarta asks that availability of the product be confirmed before conditioning begins |
| 9 Thaw and infusion | Kymriah: within 30 minutes of reaching room temperature. Breyanzi: within 2 hours of removal from storage. Carvykti: infusion completed within 2.5 hours of thawing. Yescarta: up to 3 hours at room temperature after thawing |
All Sourced (FDA prescribing information [Refs. 1 to 7]).
5. How long manufacturing takes, and manufacturing failure
The clinical studies section of the prescribing information gives the number of days from apheresis to the product arriving (or becoming available). Below is what could be confirmed in the prescribing information for each product this article consultedSourced.
| Product (trial) | Collection to product (median, range) | Collection to infusion (median) | What is said about manufacturing failure |
|---|---|---|---|
| Tecartus (MCL, NCT02601313) | Delivery 15 days (11 to 28) | 27 days | 3 of 74 patients who underwent leukapheresis (4%) |
| Yescarta (ZUMA-1, NCT02348216) | Delivery 17 days (14 to 51) | 24 days | 1 of 111. Risk under 1% |
| Aucatzyl (FELIX) | Release 20 days (17 to 23) | 35 days | Five cases mentioned |
| Breyanzi (NCT03575351) | Availability 26 days (19 to 84) | 36 days | Risk 11% |
| Carvykti (CARTITUDE-1) | Availability 32 days (27 to 66) | — | 18% (17 of 97) |
| Abecma (NCT03651128) | Availability 35 days (24 to 102) | — | 2.4% (6 of 249) |
| Kymriah | — | — | Up to 9% of manufacturing attempts |
All Sourced (FDA prescribing information [Refs. 1 to 7]). The labels variously say "delivery", "release" and "availability", and the table keeps those distinctions. "—" means no corresponding statement could be confirmed. For Breyanzi and Abecma, the figure is from one representative trial among several.
- Yescarta (ZUMA-1): 17 ÷ 24 = about 71%
- Tecartus (MCL trial): 15 ÷ 27 = about 56%
- Aucatzyl (FELIX): 20 ÷ 35 = about 57%
- Breyanzi (NCT03575351): 26 ÷ 36 = about 72%
Assumptions and limits: these are ratios of medians, not the median of individual patients' ratiosOur calculation. "To product" presumably includes shipping, and "to infusion" includes periods such as conditioning. The calculation is meant to give a sense of scale: more than half of the wait is manufacturing and shipping, but the remaining 30 to 40% passes outside the manufacturing site (our commentary).
On the reasons for manufacturing failure, the Carvykti prescribing information counts as manufacturing failures those patients who "received CARVYKTI that did not meet product release specifications or for whom data were insufficient"Sourced. The Abecma prescribing information states that of six patients, three received CAR-positive T cells that did not meet release specifications and for three the product could not be manufacturedSourced. It also states that if manufacturing fails, a second attempt may be made, and that additional treatment may be needed in the meantimeSourced.
6. A materials engineer's view (1): the materials inside the process
Reread the DESCRIPTION sections of the prescribing information with a materials eye and the CAR-T manufacturing process turns out to contain a number of materials besides the cells.
- Beads: Kymriah activates T cells with beads coated with anti-CD3/CD28 antibodiesSourced. Particles with antibodies fixed to their surface are a "functional interface" that presents proteins on a solid surface to signal to T cells. The density of antibody on the surface and the particle size presumably bear on the strength of activation, and whether the particles can be removed afterwards on the purity of the final product (our commentary)
- Viral vectors: retroviral and lentiviral vectors are raw materials manufactured separately from the CAR-T cells (our commentary). Their quality and supply are a precondition for manufacturing the CAR-T cells at all
- Cryopreservation medium: formulations containing DMSO (5 to 7.5%), albumin, dextran and so onSourced. The Kymriah prescribing information states that hypersensitivity reactions may be due to DMSO or dextran 40Sourced
- Containers: patient-specific infusion bags and vials. Yescarta is held individually in a metal cassette and stored at the site at or below −120 °C. There is a step to check the bag for cracks and other damage before thawingSourced
What is interesting is that for the same purpose, activating T cells, some products use beads while others add antibodies and IL-2 to the liquidSourced. A solid carrier delivers the stimulus across a surface, but adds a step to separate the carrier. Dissolving the reagents in the liquid removes the need for separation, but changes how freely the stimulus can be tuned. Even a single step, activation, involves a materials design choice: whether to use a carrier material (our commentary).
7. Approved products (FDA and PMDA)
Below are the CAR-T cell products listed in the FDA's list of approved products (updated 17 September 2026), with the Japanese approvals that could be confirmed in PMDA's lists of approved products (PMDA is Japan's Pharmaceuticals and Medical Devices Agency)Sourced.
| Product (non-proprietary name) | Target | First US approval (FDA) | First Japanese approval (PMDA) |
|---|---|---|---|
| Kymriah (tisagenlecleucel) | CD19 | 30 August 2017 | 26 March 2019 |
| Yescarta (axicabtagene ciloleucel) | CD19 | 18 October 2017 | 22 January 2021 |
| Tecartus (brexucabtagene autoleucel) | CD19 | 24 July 2020 | Not confirmed |
| Breyanzi (lisocabtagene maraleucel) | CD19 | 5 February 2021 | 22 March 2021 |
| Abecma (idecabtagene vicleucel) | BCMA | 26 March 2021 | 20 January 2022 |
| Carvykti (ciltacabtagene autoleucel) | BCMA | 28 February 2022 | 26 September 2022 |
| Aucatzyl (obecabtagene autoleucel) | CD19 | 8 November 2024 | Not confirmed |
US dates are the dates of the approval letters (the earliest) on each FDA product page; Japanese dates are from PMDA's list of approved products for each fiscal year. All Sourced (FDA [Refs. 8 to 12], PMDA [Refs. 14 to 17]). Later additions of indications (partial change approvals) are not included. "Not confirmed" means this article could not find the product in PMDA's lists of approved products for FY2015 to FY2026 and is Not yet confirmed. The FDA list also includes Tecelra (afamitresgene autoleucel), which introduces a T cell receptor rather than a CAR; as it is not a CAR-T product, it is left out of the table.
PMDA's lists describe all five products approved in Japan as processed human somatic cell products in which a CAR is introduced by a viral vector into "T cells derived from the patient's peripheral blood"Sourced. Yescarta uses a retroviral vector and the other four lentiviral vectorsSourced.
8. The challenge of solid tumours
All approved CAR-T cell products are for blood cancersSourced. NCI says that compared with the progress in blood cancers, development of CAR-T cell therapy for solid tumours has lagged, and explains why as followsSourced.
NCI also mentions that promising results have been reported in a small clinical trial in children and young adults with diffuse midline gliomaSourced. However, no approval of a CAR-T cell product for solid tumours could be found in the FDA's list of approved products (updated 17 September 2026)Not yet confirmed.
9. Towards allogeneic CAR-T and CAR-NK
All approved products are autologous: they are made from the patient's own T cells. That is why collection to infusion takes weeks, and why there is no substitute if manufacturing fails (Section 5). As a way around this, research is under way on "allogeneic" products made in advance from healthy donors' cells.
| Approach | What the primary sources say |
|---|---|
| Genome-edited allogeneic CAR-T (UCART19) | Two Phase 1 trials of genome-edited, donor-derived anti-CD19 CAR-T cells, positioned as a product "available for immediate clinical use". Given to 7 children and 14 adults: cytokine release syndrome in 19 (91%), grade 1 acute skin GVHD in 2 (10%), and 2 treatment-related deaths. At day 28, 14 of 21 (67%) were in complete remission or complete remission with incomplete blood count recovery |
| Cord blood-derived CAR-NK cells | A Phase 1/2 trial giving HLA-mismatched, cord blood-derived anti-CD19 CAR-NK cells to 11 patients. No cytokine release syndrome, neurotoxicity or GVHD developed. Responses in 8 of 11 (73%). The CAR-NK cells persisted at low levels for at least 12 months |
All Sourced (Benjamin et al., Lancet 2020 [Ref. 18]; Liu et al., N Engl J Med 2020 [Ref. 19]). Both are small, early-stage trials.
NK cells, like T cells, are immune cells that attack cancer cells; the paper positions CAR-NK cells as a possible way to overcome the toxicity and manufacturing complexity of CAR-T cellsSourced. Neither allogeneic CAR-T nor CAR-NK appears in the FDA's list of approved products (updated 17 September 2026), and when they will reach practical use is not yet knownNot yet confirmed.
10. A materials engineer's view (2): who can shorten the wait?
Let us reread the table in Section 5 as a process engineer would.
- A lot size of one: one manufacturing run yields product for just one patient. Improving yield (the manufacturing failure rate) pays off not in plant utilisation but in the number of patients who can be treated (our commentary)
- The raw material cannot be chosen: it is the patient's own T cells, collected in a state affected by the disease and prior treatment. The Tecartus prescribing information states that its manufacture includes a T cell enrichment step that may reduce the likelihood of circulating CD19-positive tumour cells ending up in the productSourced. On this reading, a design that absorbs raw-material variability in the process is already built in (our commentary)
- Time outside manufacturing: of the time from collection to infusion, 30 to 40% at the median was the period after the product arrived (our calculation in Section 5)Our calculation. In routine practice, one report found a median of 62 days from referral to infusion and 32 days from collection to infusion, with factors other than manufacturing, such as insurance procedures, also contributing to delaysSourced
The materials side could contribute in places such as these (our commentary).
- Activation carriers: the surface design of carriers such as antibody-coated beads, and how easily they can be removed
- Closed culture bags and tubing: parallel manufacture at one lot per patient consumes single-use plastic products in large quantities. Low extractables and low cell adsorption govern quality
- Cryopreservation media and containers: media that allow lower DMSO concentrations, bag materials that do not crack at −150 °C, and formulations that relax the post-thaw time limit (30 minutes to 3 hours)
Running hundreds of small chemical plants side by side, each for a single patient — that is one way to describe CAR-T cell manufacturing. Every wetted surface in those plants, and every material that handles freezing and thawing, has a bearing on how long patients wait and how often treatment succeeds.
11. The challenges in summary
(1) Manufacturing: time and failure
In the pivotal trials the median time from collection to product was 15 to 35 days, and stated manufacturing failure ranged from under 1% to 18%Sourced. The share of patients receiving bridging therapy while they waited was 75% in the Carvykti trial and 85% in the Abecma trialSourced.
(2) Safety
The prescribing information carries a boxed warning covering cytokine release syndrome, neurotoxicity and the occurrence of T cell malignanciesSourced. NCI names cytokine release syndrome and neurotoxicity (ICANS) as side effects of particular concernSourced.
(3) Widening the scope
Use in solid tumours is at the research stage, facing the barriers of target antigens, an immunosuppressive environment and tumour heterogeneitySourced. Allogeneic CAR-T and CAR-NK are in early clinical trialsNot yet confirmed.
- CAR-T puts into the patient's T cells an artificial receptor that joins an antibody-derived recognition part to a T cell switchSourced
- Manufacturing runs collection → enrichment → activation → viral-vector transduction → expansion → formulation and freezing → testing and release → thawing and infusionSourced
- The median time from collection to product is 15 to 35 days. Roughly 60 to 70% of the wait (56 to 72%) was the time until the product arrivedOur calculation
- Seven products are approved in the United States and five in Japan, all autologous and all for blood cancersSourced
- Solid tumours, allogeneic CAR-T and CAR-NK are at the research stageNot yet confirmed
- Materials — beads, vectors, culture vessels, cryopreservation media and cryogenic containers — govern waiting time and success rates (our commentary)
12. Glossary
- CAR (chimeric antigen receptor)
- An artificial receptor joining an antibody-derived antigen-binding part to the signalling part of a T cell.
- T cell
- A type of immune cell that finds and attacks abnormal cells.
- scFv
- Single-chain variable fragment. Only the "recognition part" of an antibody, joined into a single chain.
- Co-stimulatory domain
- The part that sends signals helping T cells activate and multiply, such as CD28 or 4-1BB.
- CD3ζ
- The part that carries the main activating signal of a T cell.
- CD19 / BCMA
- Target antigens of CAR-T cell products. CD19 is on the surface of B cells, BCMA on plasma cells and others.
- Apheresis
- A method of collecting specific blood components, such as white blood cells, by separating them from the blood.
- Viral vector
- A virus modified so that it cannot replicate, used to carry genes into cells.
- Retrovirus / lentivirus
- Types of viral vector that integrate genes into the genome.
- Lymphodepleting chemotherapy
- Conditioning chemotherapy given before CAR-T cells are infused.
- Bridging therapy
- Treatment given to hold the disease in check while the product is being manufactured.
- Cytokine release syndrome
- A systemic reaction caused when activated immune cells release large amounts of cytokines.
- Allogeneic CAR-T
- CAR-T cells made from a healthy donor's T cells. GVHD has to be dealt with.
- CAR-NK
- NK cells into which a CAR has been introduced.
13. References
- U.S. FDA "YESCARTA (axicabtagene ciloleucel) Prescribing Information", revised 07/2026 — fda.gov
- U.S. FDA "KYMRIAH (tisagenlecleucel) Prescribing Information", revised 6/2025 — fda.gov
- U.S. FDA "BREYANZI (lisocabtagene maraleucel) Prescribing Information", revised 2/2026 — fda.gov
- U.S. FDA "CARVYKTI (ciltacabtagene autoleucel) Prescribing Information", revised 10/2025 — fda.gov
- U.S. FDA "ABECMA (idecabtagene vicleucel) Prescribing Information", revised 11/2025 — fda.gov
- U.S. FDA "TECARTUS (brexucabtagene autoleucel) Prescribing Information", revised 06/2026 — fda.gov
- U.S. FDA "AUCATZYL (obecabtagene autoleucel) Prescribing Information", revised 02/2026 — fda.gov
- U.S. FDA "KYMRIAH" product page (approval letter of 30 August 2017, among others) — fda.gov
- U.S. FDA "YESCARTA" product page (approval letter of 18 October 2017, among others) — fda.gov
- U.S. FDA "TECARTUS", "BREYANZI" and "CARVYKTI" product pages (approval letters of 24 July 2020, 5 February 2021 and 28 February 2022, among others) — fda.gov (TECARTUS) / fda.gov (BREYANZI) / fda.gov (CARVYKTI)
- U.S. FDA "ABECMA" and "AUCATZYL" product pages (approval letters of 26 March 2021 and 8 November 2024, among others) — fda.gov (ABECMA) / fda.gov (AUCATZYL)
- U.S. FDA "Approved Cellular and Gene Therapy Products" (updated 17 September 2026) — fda.gov
- National Cancer Institute "CAR T Cells: Engineering Patients' Immune Cells to Treat Their Cancers", updated 26 February 2025 — cancer.gov
- Pharmaceuticals and Medical Devices Agency (PMDA) "List of products approved in FY2018 (regenerative medicine products)" (PDF, in Japanese) — pmda.go.jp
- Pharmaceuticals and Medical Devices Agency (PMDA) "List of products approved in FY2020 (regenerative medicine products)" (PDF, in Japanese) — pmda.go.jp
- Pharmaceuticals and Medical Devices Agency (PMDA) "List of products approved in FY2021 (regenerative medicine products)" (PDF, in Japanese) — pmda.go.jp
- Pharmaceuticals and Medical Devices Agency (PMDA) "List of products approved in FY2022 (regenerative medicine products)" (PDF, in Japanese) — pmda.go.jp
- Benjamin R, et al. "Genome-edited, donor-derived allogeneic anti-CD19 chimeric antigen receptor T cells in paediatric and adult B-cell acute lymphoblastic leukaemia", Lancet 396(10266):1885-1894, 2020 — doi.org
- Liu E, et al. "Use of CAR-Transduced Natural Killer Cells in CD19-Positive Lymphoid Tumors", N Engl J Med 382(6):545-553, 2020 — doi.org
- Hu B, et al. "Real-World Analysis of Barriers to Timely Administration of Chimeric Antigen Receptor T Cell (CAR T) Therapy in Diffuse Large B-cell Lymphoma", Transplant Cell Ther 30(11):1082.e1-1082.e10, 2024 — doi.org
- Neelapu SS, et al. "Axicabtagene Ciloleucel CAR T-Cell Therapy in Refractory Large B-Cell Lymphoma", N Engl J Med 377(26):2531-2544, 2017 — doi.org
14. Claim-to-source audit
| Claim in the text | Basis | Label |
|---|---|---|
| That Yescarta modifies the patient's own T cells ex vivo with a retrovirus; its CAR make-up (murine anti-CD19 scFv + CD28 + CD3ζ); the manufacturing flow (apheresis, T cell enrichment, activation with anti-CD3 antibody and IL-2, transduction, expansion, washing, formulation, freezing, shipment in a patient-specific bag after sterility testing); DMSO 5%; dry shipper at or below −150 °C and at or below −120 °C at the site; metal cassette; damage check; up to 3 hours at room temperature after thawing; matching of patient identifiers; confirming the product before conditioning; in ZUMA-1 (Study 2; NCT02348216), 17 days to delivery and 24 days to infusion; one manufacturing failure and risk under 1%; a second manufacturing attempt and treatment while waiting; and the boxed warning on T cell malignancies with lifelong monitoring | FDA YESCARTA prescribing information — Reference 1 https://www.fda.gov/media/108377/download | Sourced |
| Kymriah's CAR make-up (murine anti-CD19 scFv + CD8 hinge and transmembrane + 4-1BB + CD3ζ); lentivirus; activation with anti-CD3/CD28 antibody-coated beads; DMSO 7.5% and dextran 40; at or below −120 °C; within 30 minutes of reaching room temperature; manufacturing failure of up to 9%; and that hypersensitivity may be due to DMSO or dextran 40 | FDA KYMRIAH prescribing information — Reference 2 https://www.fda.gov/media/107296/download | Sourced |
| Breyanzi's CAR make-up (FMC63-derived scFv + IgG4 hinge + CD28 transmembrane + 4-1BB + CD3ζ); the defined composition with CD8 and CD4 components activated and transduced separately; lentivirus; 75% CryoStor CS10 (DMSO 7.5%); within 2 hours of removal from storage; in NCT03575351, 26 days to availability and 36 days to infusion; manufacturing failure risk of 11% | FDA BREYANZI prescribing information — Reference 3 https://www.fda.gov/media/145711/download | Sourced |
| Carvykti's CAR make-up (two single-domain antibodies + 4-1BB + CD3ζ); lentivirus; DMSO 5%; below −120 °C; within 2.5 hours of thawing; in CARTITUDE-1, 32 days to availability; 75% received bridging therapy; manufacturing failure of 18% (17 of 97, release specifications not met or data insufficient) | FDA CARVYKTI prescribing information — Reference 4 https://www.fda.gov/media/156560/download | Sourced |
| Abecma's CAR make-up (murine anti-BCMA scFv + CD8α hinge and transmembrane + 4-1BB + CD3ζ); lentivirus; anti-CD3 and anti-CD28 antibodies and IL-2; genomic insertion site analysis; in NCT03651128, 35 days to availability; 85% received bridging therapy; manufacturing failure of 2.4% (6 of 249, of whom 3 received product not meeting release specifications and for 3 the product could not be manufactured) | FDA ABECMA prescribing information — Reference 5 https://www.fda.gov/media/147055/download | Sourced |
| Tecartus's CAR make-up (murine anti-CD19 scFv + CD28 + CD3ζ); retrovirus; anti-CD3 and anti-CD28 antibodies and IL-2; that the T cell enrichment step may reduce the likelihood of CD19-positive tumour cells in the product; dry shipper at or below −150 °C; in the MCL trial (NCT02601313), 15 days to delivery and 27 days to infusion; manufacturing failure in 3 of 74 (4%) | FDA TECARTUS prescribing information — Reference 6 https://www.fda.gov/media/140409/download | Sourced |
| Aucatzyl's CAR make-up (murine anti-CD19 scFv + 4-1BB + CD3ζ); lentivirus; in the FELIX trial, 20 days to release and 35 days to infusion; the mention of five manufacturing failures | FDA AUCATZYL prescribing information — Reference 7 https://www.fda.gov/media/183463/download | Sourced |
| That Kymriah was first approved on 30 August 2017 | FDA KYMRIAH product page — Reference 8 https://www.fda.gov/vaccines-blood-biologics/cellular-gene-therapy-products/kymriah | Sourced |
| That Yescarta was first approved on 18 October 2017 | FDA YESCARTA product page — Reference 9 https://www.fda.gov/vaccines-blood-biologics/cellular-gene-therapy-products/yescarta | Sourced |
| That Tecartus was first approved on 24 July 2020, Breyanzi on 5 February 2021 and Carvykti on 28 February 2022 | FDA product pages — Reference 10 https://www.fda.gov/vaccines-blood-biologics/cellular-gene-therapy-products/tecartus https://www.fda.gov/vaccines-blood-biologics/cellular-gene-therapy-products/breyanzi https://www.fda.gov/vaccines-blood-biologics/cellular-gene-therapy-products/carvykti | Sourced |
| That Abecma was first approved on 26 March 2021 and Aucatzyl on 8 November 2024 | FDA product pages — Reference 11 https://www.fda.gov/vaccines-blood-biologics/abecma https://www.fda.gov/vaccines-blood-biologics/aucatzyl | Sourced |
| That seven CAR-T products are listed in the FDA's list of approved products (updated 17 September 2026); that Tecelra is listed; and that no allogeneic CAR-T, CAR-NK or solid-tumour CAR-T product is listed | FDA Approved Cellular and Gene Therapy Products — Reference 12 https://www.fda.gov/vaccines-blood-biologics/cellular-gene-therapy-products/approved-cellular-and-gene-therapy-products | Sourced |
| That a CAR spans the cell membrane, with antibody fragments outside and signalling and co-stimulatory domains inside; that all approved products are for blood cancers; the lag in solid-tumour development and its three reasons (antigens, immunosuppressive environment, tumour heterogeneity); the mention of a small trial in diffuse midline glioma; and CRS and ICANS | NCI, CAR T Cells — Reference 13 https://www.cancer.gov/about-cancer/treatment/research/car-t-cells | Sourced |
| Kymriah's approval in Japan (26 March 2019; a CAR recognising CD19 introduced by lentiviral vector into T cells from the patient's peripheral blood) | PMDA list of products approved in FY2018 — Reference 14 https://www.pmda.go.jp/files/000231456.pdf | Sourced |
| The Japanese approvals of Yescarta (22 January 2021, retroviral vector) and Breyanzi (22 March 2021, lentiviral vector) | PMDA list of products approved in FY2020 — Reference 15 https://www.pmda.go.jp/files/000240837.pdf | Sourced |
| Abecma's approval in Japan (20 January 2022, lentiviral vector, BCMA) | PMDA list of products approved in FY2021 — Reference 16 https://www.pmda.go.jp/files/000246141.pdf | Sourced |
| Carvykti's approval in Japan (26 September 2022, lentiviral vector, BCMA) | PMDA list of products approved in FY2022 — Reference 17 https://www.pmda.go.jp/files/000252078.pdf | Sourced |
| The two Phase 1 trials of UCART19 (7 children and 14 adults; CRS 91%; grade 1 acute skin GVHD in 2; 2 treatment-related deaths; 67% in complete remission or similar at day 28; positioned as a product available for immediate clinical use) | Benjamin et al., Lancet 2020 — Reference 18 https://doi.org/10.1016/S0140-6736(20)32334-5 | Sourced |
| The Phase 1/2 trial of HLA-mismatched cord blood-derived CAR-NK cells (11 patients; no CRS, neurotoxicity or GVHD; responses in 8; persistence for 12 months or more); and their positioning as a possible way to overcome the toxicity and manufacturing complexity of CAR-T cells | Liu et al., N Engl J Med 2020 — Reference 19 https://doi.org/10.1056/NEJMoa1910607 | Sourced |
| In routine practice, a median of 62 days from referral to infusion and 32 days from collection to infusion, with insurance procedures contributing to delays | Hu et al., Transplant Cell Ther 2024 — Reference 20 https://doi.org/10.1016/j.jtct.2024.09.007 | Sourced |
| That ZUMA-1 is NCT02348216, and that manufacturing succeeded for 110 of 111 patients | Neelapu et al., N Engl J Med 2017 — Reference 21 https://doi.org/10.1056/NEJMoa1707447 | Sourced |
| The ratios "to product ÷ to infusion" (about 71%, 56%, 57% and 72%) and the framing that 30 to 40% of the wait comes after the product arrives | Our calculation. Ratios of medians, not individual patients' values | Our calculation |
| Japanese approval of Tecartus and Aucatzyl; CAR-T for solid tumours; approval and timing of practical use of allogeneic CAR-T and CAR-NK | Could not be confirmed within this article's research (PMDA lists of approved products for FY2015 to FY2026 and the FDA list of approved products) | Not yet confirmed |
| Kymriah's manufacturing time; days to infusion for Carvykti and Abecma; details of the activation reagents (Carvykti, Aucatzyl) | Not stated because no corresponding statement could be confirmed in the prescribing information this article consulted (commentary) | Commentary |
| Dividing the CAR into parts 1 to 5, and summarising the manufacturing process in nine steps; the explanation that genomic integration means the CAR is passed on after division; the reading of beads as a functional interface, that vectors are separately manufactured raw materials, and the materials choice of whether to use a carrier; the reading that improving the failure rate pays off in the number of patients treated, and the design that absorbs raw-material variability in the process; and the summary of where the materials side could contribute | This article's summary and commentary based on published content. Not views expressed by the companies or institutions | Commentary |
| That Figs. 1 to 5 are explanatory drawings rather than real molecular structures or manufacturing instructions, and that the hero image is an AI-generated image | Our note | Commentary |
Last updated 23 September 2026. Sources are limited to primary material (FDA prescribing information, product pages and list of approved products; PMDA lists of approved products; NCI; and peer-reviewed papers). Because the article includes structural summaries and readings about materials and processes, those are marked as Commentary and kept separate from sourced fact. Clinical trial figures are given only as far as the prescribing information and papers state them, and are not a comparison of therapeutic effect between products or a recommendation. Kymriah's manufacturing time, Japanese approval of Tecartus and Aucatzyl, CAR-T for solid tumours, and when allogeneic CAR-T and CAR-NK will reach practical use are not stated here because no published primary source could be confirmed. All figures are explanatory concept graphics. Figs. 1 to 5 are vector drawings and the hero image is an AI-generated image; none of them shows a real molecular structure, manufacturing facility or product.